OpenAI deprecated o3-pro: shuts down Dec 11, 2026 (move to gpt-5.6-sol)
What to do: Move to gpt-5.6-sol before Dec 11, 2026.
AIWire: AI model shutdowns, prices and API changes
The wire
Every change we detect, newest first. Times are Eastern (ET). Archive
TodayOctober 23, 2026
Elevated errors for Claude Fable 5.1, Sonnet 5.5, and Haiku 5.5
Heads-upAnthropic
What to do: No action; resolved.
Anthropic Python SDK 2.3.0
What to do: Update to anthropic Python SDK 2.3.0 if you use Claude Tag endpoints or Google Cloud integrations.
OpenAI deprecated o3: shuts down Dec 11, 2026 (move to gpt-5.6-sol)
What to do: Move to gpt-5.6-sol before Dec 11, 2026.
OpenAI deprecated gpt-5-pro: shuts down Dec 11, 2026 (move to gpt-5.6-sol)
What to do: Move to gpt-5.6-sol before Dec 11, 2026.
OpenAI deprecated gpt-5-nano: shuts down Dec 11, 2026 (move to gpt-5.6-luna)
What to do: Move to gpt-5.6-luna before Dec 11, 2026.
OpenAI deprecated gpt-5-mini: shuts down Dec 11, 2026 (move to gpt-5.6-terra)
What to do: Move to gpt-5.6-terra before Dec 11, 2026.
OpenAI deprecated gpt-5: shuts down Dec 11, 2026 (move to gpt-5.6-sol)
What to do: Move to gpt-5.6-sol before Dec 11, 2026.
OpenAI deprecated gpt-4o-mini-tts: shuts down Jan 6, 2027 (move to gpt-realtime-2.1-mini)
What to do: Move to gpt-realtime-2.1-mini before Jan 6, 2027.
xAI Python SDK 1.21.0
What to do: Check the xAI Python SDK changelog or repository for details before upgrading to v1.21.0.
Kimi K3.1 scores 55 on Artificial Analysis, placed on the Pareto line
InfoReleasesHacker News · community
What to do: Developers considering the model should weigh its above-average intelligence against slower output speed and high verbosity when estimating latency and token costs.
Moonshot AI added kimi-k3.1-pro-highspeed
What to do: Call it as kimi-k3.1-pro-highspeed.
Kimi K3.1 Pro and K3.1 Pro Highspeed
What to do: Use exact model IDs like k3.1-pro or k3.1-pro-highspeed, keep thinking enabled, and start new sessions when switching models or effort levels.
NVIDIA added nemotron-3.5-bolt-e4b-4gb
What to do: Call it as nemotron-3.5-bolt-e4b-4gb.
Anthropic Python SDK 2.2.0
What to do: Update to SDK 2.2.0 and review the new usage/cost report endpoints if you rely on the legacy reports.
ft-davinci-002 has been retired
What to do: Requests to ft-davinci-002 now fail. Switch to gpt-5.6-terra.
ft-babbage-002 has been retired
What to do: Requests to ft-babbage-002 now fail. Switch to gpt-5.6-terra.
ft-gpt-4.1-nano-2025-04-14 has been retired
What to do: Requests to ft-gpt-4.1-nano-2025-04-14 now fail. Switch to gpt-5.6-luna.
ft-gpt-4 has been retired
What to do: Requests to ft-gpt-4 now fail. Switch to gpt-5.6-sol.
ft-gpt-3.5-turbo has been retired
What to do: Requests to ft-gpt-3.5-turbo now fail. Switch to gpt-5.6-terra.
o4-mini-2025-04-16 has been retired
What to do: Requests to o4-mini-2025-04-16 now fail. Switch to gpt-5.6-terra.
ft-o4-mini-2025-04-16 has been retired
What to do: Requests to ft-o4-mini-2025-04-16 now fail. Switch to gpt-5.6-terra.
o3-mini-2025-01-31 has been retired
What to do: Requests to o3-mini-2025-01-31 now fail. Switch to gpt-5.6-sol.
o1-pro-2025-03-19 has been retired
What to do: Requests to o1-pro-2025-03-19 now fail. Switch to gpt-5.6-sol.
o1-2024-12-17 has been retired
What to do: Requests to o1-2024-12-17 now fail. Switch to gpt-5.6-sol.
gpt-image-1 has been retired
What to do: Requests to gpt-image-1 now fail. Switch to gpt-image-2.5-sunburst.
gpt-4o-2024-05-13 has been retired
What to do: Requests to gpt-4o-2024-05-13 now fail. Switch to gpt-5.6-sol.
gpt-4.1-nano has been retired
What to do: Requests to gpt-4.1-nano now fail. Switch to gpt-5.6-luna.